Correctional Lens Doublets consist of two lenses generally made of different optical materials attached to each other to derive better optical performance over wavelengths of interest. Currently, the most popular methods of achieving micron level centration of the two pieces of lens doublets involve active alignment or other methods. Unfortunately, these methods are slow, high in cost and not amenable to volume production. The problem of centration is even more difficult for very small lens pieces such as used in handheld cameras or cell phone cameras.
Active alignment involves passing an optical beam through the lenses or lens assemblies. Controlled movement of the lenses is executed using the location of the projected beam image for feedback. The Laser Alignment and Assembly Station™ of Opto-Alignment Technology, Inc. is an example of a apparatus that utilizes active alignment. A rotating air bearing stage has X-Y-theta-phi adjustments to set the lens barrel or other optic holder to run true to serve as a reference. Then a reflection from a laser imaged on a CCD camera is used to center multiple optical elements.
There are several ways in which the lens can be centered on the bell-shaped centering bell. Many of these methods are discussed with figures in “Fabrication Methods for Precision Optics,” by Hank H. Karow. Brief descriptions are provided here.
An older transfer spindle method is still used for prototype and low-volume production and for centering lenses made from soft and sensitive materials. According to Karow, the lens is mounted on a precise centering bell which is fastened on a thread to a precision spindle. This centering bell is nearly always made of brass, although steel bells are sometimes used as well. The lens is mechanically or optically aligned on the centering bell so that the optical and mechanical axes of the lens are coincident. The spindle with the centered lens is then mounted in a centering machine, which is used to grind the diameter of the lens concentric to the axis of rotation.
In one centering method, the lens and the centering bell are heated sufficiently so that a small amount of a specially formulated centering wax can be applied to the edge of the centering bell. The heated lens is then positioned on the centering bell by mechanical means or by an optical alignment method. For the mechanical method, the edge thickness variations must be nulled out or at least reduced to a minimum while the lens rotates. The run-out is monitored by means of a digital indicator.
Another centering method is an optical method that relies on light reflected off the outer surface of the lens. The lens is mounted on a centering bell, which rotates the lens. The centering bell is mounted to the spindle.
Yet another centering method is an optical method that relies on the rotation of a target image that is projected through the lens. A lamp, a collimating lens and mirrors form the target image and project the target image through the lens. The lens is mounted on the centering bell, which rotates the lens.
A laser beam alignment method is used for more critical centration requirements. It is quite similar to the target image projection method. Both of these optical centration methods are only useful for lenses made from visually transparent materials. Projection methods are not possible for infrared materials and metal optics because they are not transparent in the visible region of the spectrum.
Another centering method is bell chuck centering for small lenses. Using two coaxial bell chucks to align a single lens to the chuck axis. The lens is typically arranged horizontally. Bell chuck centering is a method in which a suitably curved lens is self-centering when clamped between identical and precision aligned brass centering bell or bell chucks. Bell chuck centering is a purely mechanical method that relies on the fact that the lens will slide along the edge of the bell chuck until it seats itself at the zone of equal edge thickness. In this alignment the optical axis of the lens and the mechanical axis of the spindle become colinear. When the edges of the centering chucks are well rounded and polished, the lens surface will not be damaged, even at high chuck pressures. The bell chuck centering method with a horizontal spindle can be used for lenses with diameters as small as 3 mm and as large as 150 mm.
Still another centering method is bell chuck centering for larger lenses. Lenses up to 250 mm in diameter can be centered with this bell chucking method using vertically aligned spindles. For larger lenses, a vacuum assist method is necessary to reduce the otherwise required high clamping loads on the lens surfaces that can lead to damage.
Regarding centerability limitations, for the centering method known as transfer spindle, there is no theoretical optical limit on centerability, although some practical limits do exist. Even lenses with very weak optical power can be centered with this method as long as the centration error can be optically or mechanically detected. There is a well-defined centerability limit, however for the bell-chucking method. When the lens radii become too long, even strong forces will not prevent the lens from being displaced by the appreciable side pressures exerted by a side wheel typically formed of diamond. Before a lens production run is committed to a bell-chucking machine, it must first be determined if the lens can be safely, accurately and reproducibly centered this way.
The limit of centerability with the bell-chucking method is a function of the slide angle of the lens relative to the chuck edge and the coefficient of friction between lens and chuck. The motion component which can be derived from chuck pressure causes the lens to move between the bell chuck edges. This motion component must be large enough so that it can overcome the friction between the bell chuck edges and the lens surfaces. Since the value of this component depends on the shape of the lens, it is possible for it to become smaller than the friction component. The lens will then no longer slide between the bell chuck edges, and it can no longer be centered this way. The limit of centerability can be calculated from the lens diameter, the lens radii, and the coefficient of friction.
Doublet alignment is yet another prior art method based on aligning the doublet to equal lens edge thickness. The doublet includes a convex and a concave lense joined together. The two lenses of a cemented or joined doublet have a common optical axis.
For noncritical doublets it is sufficient to align the lenses by their common diameter. The doublet is put into a V-jig for this purpose. The cemented doublet is then centered as accurately as individual lenses. The centering tolerances can be additive, however.
A somewhat more accurate centering method for doublets uses a fixture in which the doublet is held in a lens holder. Mounted above the doublet is a centering bell on the same axis as the lens holder. This centering bell is lowered until it contacts the doublet, which will then automatically align the lense to an equal edge thickness. Alignment to equal edge thickness is also the principle on which an older instrument was based.
Doublets for high quality objectives are optically aligned. A transmitted light method is well known for which the setup is equipped with a collimator and a telescope. Systems are also used in which the light from an autocollimator passes through a doublet and is reflected back by a plane mirror.
Doublets can also be cemented and aligned to their common optical axis on a laser centering test unit. The cement is applied and evenly distributed. This method works best with UV cement. The lens pair is then placed on the centering chuck of the instrument where it is held by light vacuum pressure. The lower, typically concave, lens is pushed against a laterally attached V-stop which has been previously adjusted with a centered lens and locked into position.
The laser passes through a focusing apparatus that compensates for the refractive powers of the doublet. A plane mirror redirects the beam from there through the doublet, which then passes through the hollow spindle and impinges as a light spot on a detector plate. The detector is electronically coupled to an image screen that displays the light spot as a bright dot. A change in position of the light spot on the detector plane is seen magnified many times on the image screen. The upper lens is then shifted in such a way that the dot on the screen remains stationary as the spindle is rotated. The centered doublet is now irradiated with UV light The cement sufficiently precures in a few seconds, so then the cemented doublet can be removed from the centering chuck and fully cured later in a ultra violet light oven.
The Laser Alignment and Assembly Station™ by Opto-Alignment Technology, Inc., is an instrument to improve quality of assembly and inspection of precision and ultra-precision multi-element lens assemblies. The Laser Alignment and Assembly Station™ operates by reflecting a laser beam from any surface of an object being aligned or measured. A centering and tilt stage is mounted on an air bearing to bring the mechanical axis of a lens housing into collinear alignment with the incident test beam. The optical module provides a focused laser test beam which, when reflected from the surface under test, falls on a charge-coupled device (CCD) detector. If the surface is accurately centered, the reflection produces a stationary bright spot in the center of the video monitor. A misaligned surface will reflect the laser beam at an angle and displace the bright spot away from the center of the CCD. Rotation of the air bearing causes the decentered spot to orbit around the center of the CCD. The radius of the orbit viewed on a monitor is proportional to the tilt of the surface under test and is independent of its radius of curvature. A frame-grabber can be used to import the image to a computer and software can automate the measurements and store the data on the individual lenses and the entire lens assembly.
Voice coil linear actuators with flexural bearings are also part of prior art. The LFA-10 Linear Focus Actuator™ of Equipment Solutions Inc. is a positioning apparatus using such voice coil linear actuators. It was specifically developed for optical applications requiring both high precision and high-speed positioning over a short to medium stroke. The LFA-10 Linear Focus Actuator™ is well suited for optical focusing and other micropositioning applications such as scanning interferometry, surface structure analysis, disk drive testing, autofocus systems, confocal microscopy, biotechnology and semiconductor test equipment. The LFA-10 Linear Focus Actuator™ is guided along a single axis by a flexure design. The use of flexures within the design produces a compact and light package with zero stiction/friction, ultra-high resolution and exceptional guiding precision. This stage architecture allows it to be oriented in either a vertical or horizontal position. The LFA-10 Linear Focus Actuato™ uses a high force low mass moving voice coil architecture. The LFA-10 Linear Focus Actuator™ includes a sub-micron resolution linear displacement sensor.
All of the aforementioned methods are too slow to justify their capital cost of equipment and operator labor. Note that executing a lens alignment method using automated pick and place machines is possible. However, such a method is not accurate enough due to imperfections in lens outlines and poor correlation between the outline of the lens and its optical axis.